Daubert — If Expert Failed to Follow Own Stated Methods, No Reliable Method Followed — Mere Expertise & Subjective Understanding ≠ Reliable Scientific Evidence (Good Quote) — De Novo Issue before Abuse of Discretion Applies
Brown v. Burlington N. Santa Fe Rwy., 2014 U.S. App. LEXIS 16865 (7th Cir. Aug. 29, 2014):
Shannon Brown appeals the dismissal of his lawsuit against the Burlington Northern Santa Fe Railway Company ("BNSF"), which he filed under the Federal Employers' Liability Act ("FELA"), 45 U.S.C. § 51 et seq. The sole issue he disputes on appeal is the district court's1 decision to exclude the testimony of his expert witness, David Fletcher, M.D. We conclude that the district court did not abuse its discretion and therefore affirm its grant of summary judgment.
Discovery commenced, and Brown retained Dr. Fletcher to examine him and provide expert medical testimony. Dr. Fletcher's expertise in diagnosing railway work injuries and identifying their cause is unchallenged. He is licensed to practice medicine in Illinois, and is a full-time physician. He graduated from Rush Medical College in Chicago and holds [*5] a Master's Degree in Public Health from the University of California, Berkeley. Dr. Fletcher is a Fellow of the American College of Occupational and Environmental Medicine and has been appointed Clinical Assistant Professor at the University of Illinois and Southern Illinois University. In 2012 he was one of two doctors chosen to serve on the Illinois Workers' Compensation Commission. He is also the medical director of SafeWorks, Illinois, a private occupational health clinic. Starting in 1985 and continuing through his 2012 deposition, Dr. Fletcher occasionally served as an independent contractor with two railroad companies, the Norfolk Southern Corporation and the Canadian National Railway Company. In that capacity he treated work-related injuries and performed physicals, tested employees' fitness for duty, and conducted some ergonomic evaluations. He has served as an expert witness in past FELA cases.
A district court's decision to exclude expert testimony is governed by Federal Rules of Evidence 702 and 703, as construed by the Supreme Court in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). Rule 702(c) requires that an expert's testimony be "the product of reliable principles and methods." Similarly, [*15] Rule 703 requires the expert to rely on "facts or data," as opposed to subjective impressions. Daubert laid out four factors by which courts can evaluate the reliability of expert testimony: (1) whether the expert's conclusions are falsifiable; (2) whether the expert's method has been subject to peer review; (3) whether there is a known error rate associated with the technique; and (4) whether the method is generally accepted in the relevant scientific community. 509 U.S. at 593-94.
Dr. Fletcher sought to offer a differential etiology in this case. "Differential diagnosis is an accepted and valid methodology for an expert to render an opinion about the identity of a specific ailment." Myers, 629 F.3d at 644. So is differential etiology, which focuses on the cause, not just the identity, of an ailment. Id. But an expert still must faithfully apply the method to the facts at hand. A differential etiology, like a differential diagnosis, "satisfies a Daubert analysis if the expert uses reliable methods. ... Determining the reliability of an expert's differential diagnosis is a case-by-case determination." Ervin v. Johnson & Johnson, Inc., 492 F.3d 901, 904 (7th Cir. 2007); see also Myers, 629 F.3d at 644. The party offering the expert testimony bears the burden of proving its reliability. Lewis v. CITGO Petroleum Corp., 561 F.3d 698, 705 (7th Cir. 2009).
In reviewing the district court's [*16] decision to exclude expert testimony, this court "first undertakes a de novo review of whether the district court properly followed the framework set forth in Daubert ... ." United States v. Hall, 165 F.3d 1095, 1101 (7th Cir. 1999). If the court properly understood its role therein, we then review its ultimate decision to exclude expert testimony for an abuse of discretion. Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999). This deference is in keeping with the district court's vital "gatekeeping" role in ensuring that only helpful, legitimate expert testimony reaches the jury. Daubert, 509 U.S. at 597.
Brown first contends that the district court "exceeded the scope of its gatekeeping function" under Daubert by nitpicking Dr. Fletcher's factual observations and gainsaying his conclusions--both of which are properly roles for the jury--rather than simply determining whether he used a reliable method. Appellant's Br. at 36. Second, Brown argues that Dr. Fletcher did properly adhere to his method of differential etiology and that the district court abused its discretion in finding otherwise. We take both arguments in turn.
The district court properly understood the Daubert framework. It noted that "[t]he court's role as gatekeeper is strictly limited to an examination of the expert's methodology." Brown v. Burlington N. Santa Fe Ry. Co., No. 09-1380, 2013 WL 1729046 at *8 (C.D. Ill. Apr. 22, 2013). Brown responds that [*17] the district court's reasoning belies that acknowledgement. Specifically, the district court faulted Dr. Fletcher for apparently relying on Brown's recitation of his medical history to conclude that he did not smoke when it appears that he did. The court also noted that Dr. Fletcher's report did not describe Brown's family history of cumulative trauma disorder. Brown's failure to accurately relay his medical history should not have led the court to impugn Dr. Fletcher's methods. "Medical professionals reasonably may be expected to rely on self-reported patient histories." Walker v. Soo Line R.R. Co., 208 F.3d 581, 586 (7th Cir. 2000) (citing Cooper v. Carl A. Nelson & Co., 211 F.3d 1008, 1019-21 (7th Cir. 2000)). Likewise, the district court doubted Brown's self-reporting of his job duties to Dr. Fletcher, in particular his claims that he worked 12-16 hours a day. The court also discounted Dr. Fletcher's diagnosis of Brown's shoulder injury because he was not able to perform an MRI. Finally, the district court accused Dr. Fletcher of misidentifying a track jack as an iron angle bar, which would be a significant error because the two items are quite distinct. Brown argues that the district court's findings on these points amount to improperly quibbling with factual details of the expert's report. After all, even [*18] experts make mistakes, and imperfections in their presentations are supposed to be tested by opposing counsel and put before the jury.
Although the district court did observe factual deficiencies in Dr. Fletcher's reports, it clearly stated that it was excluding the doctor's testimony because he failed to follow a reliable method; indeed, he deviated from his own stated description of a job site analysis and of differential etiology in general. Dr. Fletcher entirely failed to personally observe Brown's working conditions, obtain a written work description, or perform scientific tests. He also failed to investigate several possible causes of Brown's health problems. The factual deficiencies or discrepancies the district court identified are the result of Dr. Fletcher's faulty methods and lack of investigation. The district court used the gaps in Dr. Fletcher's analysis as illustrative examples of the perils inherent in applying subjective experience instead of a proper scientific approach. The district court did not exceed its role under Daubert.
Brown's remaining argument is that the district court abused its discretion in finding that Dr. Fletcher failed to apply a reliable method. [*19] We have recognized that there is "nothing controversial" about using differential etiology to establish legal cause. Schultz v. Akzo Nobel Paints, LLC, 721 F.3d 426, 433 (7th Cir. 2013). However, an expert must do more than just state that she is applying a respected methodology; she must follow through with it. In deciding whether an expert employed a reliable method, the district court has discretion to consider "'[w]hether the expert has adequately accounted for obvious alternative explanations.'" Id. at 434 (quoting Fed. R. Evid. 702 (2000) Committee Note). The expert need not exclude all alternatives with certainty, however. See Gayton v. McCoy, 593 F.3d 610, 619 (7th Cir. 2010) ("[A]n expert need not testify with complete certainty about the cause of an injury; rather he may testify that one factor could have been a contributing factor to a given outcome.").
The district court did not abuse its discretion in finding that Brown's motorcycle riding and volunteer firefighting were obvious potential alternative causes for his injuries. The causal link Dr. Fletcher drew between Brown's job and his injuries lay in the presence of vibratory and other types of equipment that can harm elbows, wrists, and shoulders over time. But the handlebars of a running motorcycle obviously vibrate, and firefighters must frequently struggle with heavy equipment. [*20] Brown is correct that under FELA he need only prove that BNSF's negligence was a cause, not the sole cause, of his injury. But without performing an investigation, Dr. Fletcher could not rule out either activity as the sole cause of Brown's condition. And although Brown's weight, history of smoking, and family medical history were each not likely the sole cause of his ailments, these risk factors combined with either the volunteer firefighting or motorcycle riding (or both) could have been wholly responsible for Brown's condition. We do not know how likely this possibility is because Dr. Fletcher did not meaningfully consider it.
Brown insists that Dr. Fletcher did consider these obvious alternative causes, but the record shows otherwise. The doctor disregarded Brown's motorcycle riding as a factor because he assumed Brown worked for longer periods than he rode. But as BNSF rightly points out, the proper question is how long he rode the motorcycle as compared to how long he used vibratory or similarly taxing tools at work. And Dr. Fletcher could not possibly answer that question in a systematic, testable fashion because he did not know the duration and frequency of Brown's motorcycle [*21] riding. Even worse, he did not know the duration or frequency of Brown's exposure to vibrations at work. He did not have enough information to conclude that one value was higher than the other, or even to doubt that the former overwhelmingly exceeded the latter. Comparing two unknown, potentially wide-ranging variables is not a scientific exercise. There is no known error rate attached to such a calculation, nor is such guesswork widely accepted in the scientific community. See Daubert, 509 U.S. at 593-94. Similarly, Dr. Fletcher did not know what hours Brown worked as a firefighter, or what his responsibilities were. These were not merely factual oversights; they are flatly inconsistent with differential etiology. That method does not establish a cause for an injury directly, through observation or factual reconstruction. Rather, it relies on the process of elimination by ruling out other alternatives. The failure to rule out obvious potential alternative causes is therefore fatal to Dr. Fletcher's testimony.
Dr. Fletcher's failure to consider Brown's motorcycle riding and volunteer firefighting distinguishes this case from Schultz, 721 F.3d 426, which Brown cites in support of his argument. In that case, the plaintiff had smoked [*22] in the past, but the expert explicitly stated that exposure to benzene was known to pose an even greater risk. This meant, in the expert's opinion, that the benzene was a "substantial factor" in his cancer. Id. at 433-34. Here, Dr. Fletcher did not reliably weigh the risks posed by Brown's job-related exertion as compared to his other activities. This case is also quite different from Hardyman, in which the expert "took an extensive history of Plaintiff's non-occupational work activities." 243 F.3d at 261 (discussing the plaintiff's bowling, golf, and other recreational activities).
Not only did Dr. Fletcher fail to investigate and systematically rule out two obvious potential causes, but it is not clear that he ruled out any serious alternative. It is true that Brown apparently does not have diabetes, which could be a risk factor. Dr. Fletcher also determined that Brown's weight was not likely not a problem, because his BMI was "[b]orderline." But even this is difficult to square with his observation during his deposition that "[t]he higher the BMI the more likely that obesity could be an independent risk factor" for carpal tunnel syndrome. He did not explain at all why this positive relationship would exist only for the "morbidly [*23] obese." Brown's weight could have made it more likely that his motorcycle riding or volunteer firefighting was solely responsible for his condition. Of course, we can only speculate because Dr. Fletcher did not adequately investigate this possibility.
As the district court correctly observed, Dr. Fletcher's failure to rule out obvious potential causes was only half the problem. He also failed to reliably "rule in" Brown's workplace activity as a potential cause of Brown's condition. Dr. Fletcher failed to consider that Brown's job duties changed considerably as he progressed, beginning in 2006, from maintenance-of-way work to different roles as a foreman, track inspector, and machine operator. More fundamentally, Dr. Fletcher noted that his method required him to conduct a "job site analysis." This involved "traveling to the literal worksite with the patient and reviewing his or her job duties; measuring frequency and force required for various job tasks; videotaping and photographing job task activities for further analysis"; identifying "variances in the written job description as compared to the actual duties performed; using scientific measuring tools, such as a Chatillon gauge, [*24] which constitutes an objective measure of force; assessing push/pull job function factors; and evaluating the level of force exertion required to perform a job task." Observing Brown's actual working conditions was important in order to avoid "ruling in" risk factors that were not actually present at his job. The use of videotape and photography to record Dr. Fletcher's observations would have been crucial to ensuring that his conclusions could be objectively tested, peer reviewed, and reproduced. The same applies to the use of scientific tools that provide recorded measurements and the written job description that could offer an objective comparison with the doctor's observation. Dr. Fletcher also testified that he usually had a professional ergonomist conduct much of this investigation, but he did not use his services in this case. All of these steps are designed so that the expert can rely not on his own subjective experience or bias but on reliable scientific methods. Dr. Fletcher noted that the above safeguards were important in his own report, yet he failed to follow them. This again distinguishes Brown's case from Hardyman, where the plaintiff's ergonomics expert "conducted [*25] an extensive investigation of Plaintiff's work conditions." 243 F.3d at 263. Without a legitimate investigation, Dr. Fletcher could not reliably ascertain whether Brown's work was even a contributing factor to his injury.
In response, Brown contends that precise measurements of the duration and frequency of his exposure to vibratory and other potentially damaging tools are unnecessary because no precise relationship between the frequency and duration of exposure and a particular cumulative trauma injury is known. Indeed, it likely varies from patient to patient. But because Brown was exposed to multiple sources of continued vibration and other trauma, Dr. Fletcher had to have some reliable basis for opining that Brown's work activities played at least a small role in his injury. Data comparing the relative duration and frequency of exposure could have provided that basis; perhaps there were other ways. But Dr. Fletcher did not pursue any of them. Brown also argues that BNSF's experts also did not perform frequency and duration tests of its equipment either, but pointing out deficiencies in the defendant's expert testimony cannot help Brown, who bears the burden of proving negligence and demonstrating [*26] the reliability of his own expert.
Brown claims that Dr. Fletcher was prevented from conducting the type of job site analysis described in his reports because BNSF would not cooperate by, for example, allowing him to test its tools or providing him with a written job description. But that is a matter that should have been brought to the district court's attention during discovery. A party cannot enter into evidence unhelpful expert testimony on the grounds that the other side made them do it. If Brown felt that BNSF was unreasonably constraining his expert's investigation, he should have raised that issue and then, if unsuccessful, pressed it on appeal.
Nor did Dr. Fletcher follow his own advice in diagnosing Brown's alleged shoulder injury. In his first report he noted that a "formal diagnosis" would not be possible without an MRI. The district court did not abuse its discretion in holding Dr. Fletcher to that representation. And if Dr. Fletcher failed to follow his own stated methods, the court could reasonably conclude that he had failed to follow any reliable method. Brown has not shown that Dr. Fletcher's actual approach, as opposed to what he claimed to have done, was generally [*27] accepted in the scientific community. His process could not produce falsifiable results or survive peer review, and it is impossible to put an error rate on his guesswork. See Daubert, 509 U.S. at 593-94.
No one disputes that Brown's injuries could have been caused by frequent or long-lasting vibrations, or that his job exposed him to a significant amount of vibration over the years. But if that were sufficient to establish causation, expert testimony would be unnecessary in this case. Any layman can understand that connection. Brown wishes to use Dr. Fletcher's quarter-century of experience in the field to rule out other potential causes. But experience without reliable, testable methodology is not sufficient. See Gen. Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997) ("[N]othing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert."). Moreover, Dr. Fletcher's application of his own experience is itself suspect. At his deposition he was forced to rely on his memory of "spen[ding] half an hour, 40 minutes ... a decade ago" at a different rail yard to describe the type of maintenance-of-way work that Brown performed. The vagueness of this testimony is a good illustration [*28] of why mere expertise and subjective understanding are not reliable scientific evidence. The district court did not abuse its discretion by concluding that opinions based on this sort of recollection would be no help to the jury.
Share this article: